[MRG] DOC add an example for permutation-importance#16223
[MRG] DOC add an example for permutation-importance#16223magda-zielinska wants to merge 5 commits intoscikit-learn:masterfrom
Conversation
| Number of Monte Carlo samples per original feature. | ||
| Equals the dimensionality of the computed feature space. | ||
|
|
||
| random_state : int, RandomState instance or None, optional (default=None) |
There was a problem hiding this comment.
Could you revert all the changes in the kernel_approximation file. It is not related with your PR.
| LogisticRegression() | ||
|
|
||
| >>> result = permutation_importance(clf, X, y, n_repeats=10, | ||
| ... random_state=42) |
There was a problem hiding this comment.
| ... random_state=42) | |
| ... random_state=42) |
|
|
||
| >>> clf = LogisticRegression() | ||
| >>> clf.fit(X,y) | ||
| LogisticRegression() |
There was a problem hiding this comment.
| LogisticRegression() | |
| LogisticRegression(...) |
| ... [0,9,9],[0,9,9],[0,9,9]] | ||
| >>> y = [1,1,1,0,0,0] | ||
|
|
||
| >>> clf = LogisticRegression() |
There was a problem hiding this comment.
| >>> clf = LogisticRegression() | |
| >>> clf = LogisticRegression().fit(X, y) |
| >>> y = [1,1,1,0,0,0] | ||
|
|
||
| >>> clf = LogisticRegression() | ||
| >>> clf.fit(X,y) |
|
|
||
| >>> result = permutation_importance(clf, X, y, n_repeats=10, | ||
| ... random_state=42) | ||
|
|
There was a problem hiding this comment.
could you print results before to check a key of the bunch
| >>> clf = LogisticRegression() | ||
| >>> clf.fit(X,y) | ||
| LogisticRegression() | ||
|
|
There was a problem hiding this comment.
remove the empty line as well
| >>> clf.fit(X,y) | ||
| LogisticRegression() | ||
|
|
||
| >>> result = permutation_importance(clf, X, y, n_repeats=10, |
There was a problem hiding this comment.
| >>> result = permutation_importance(clf, X, y, n_repeats=10, | |
| >>> result = permutation_importance(clf, X, y, n_repeats=2, |
There was a problem hiding this comment.
We want to have an execution time very small so 2 will be enough.
|
Hi @magda-zielinska , could you find some time to address the reviewer comments? Thanks for participating to the sprint! |
|
Hello @cmarmo, I will take care of it in the next days! Thanks a lot! |
|
gentle ping @magda-zielinska |
|
good mornig everybody, apologies that this took so long! |
|
Do not hesitate to ping |
Together with @fraboeni and @lopusz
Reference Issues/PRs
#3846
What does this implement/fix? Explain your changes.
Added an example to permutation importance
@wimlds
@adrinjalali
@noatamir